The goal of this study is to test if a two-way text-message (SMS) maternally administered malnutrition monitoring system (MAMMS) that delivers infant and young child feeding (IYCF) education and supports caregivers in monitoring their child's nutritional status at home can improve nutritional outcomes for HIV-exposed children.

The aims include 1) to determine whether the MAMMS IYCF intervention lowers the incidence of malnutrition, leads to a shorter time to recover for those that become malnourished and results in a lower incidence of hospitalizations, severe malnutrition and death, 2) to determine the cost and cost-effectiveness of the MAMMS IYCF intervention, and 3) to determine the effect of the MAMMS IYCF intervention on the behavior and attitudes of participants through change in age-appropriate feeding, IYCF knowledge, trust in the healthcare system, and intention to seek care if the child becomes wasted.

The study team will enroll 776 caregiver-child pairs aged between 6 and 24 months in Migori and Homa Bay County, Kenya. Each caregiver-child pair will be randomly assigned to either the MAMMS IYCF intervention or standard of care (SOC) and followed for 180 days (about 6 months).

Caregivers assigned to the intervention arm will be asked to respond to weekly messages with the color of the MUAC tape after measuring their child's arm after being trained on how to use the MUAC measuring tape. Weekly messages will include IYCF education and other age-appropriate child health related information. Caregivers in the SOC arm will receive clinic appointment and study visit reminders only. Caregivers in the intervention arm and the SOC arm will be asked to attend the study clinic for follow-up visits at Day 90 and Day 180. At enrollment and follow-up visits, the study team will administer a survey including a child's medical history, a standardized child clinical examination, and anthropometry.

The investigators propose an individually randomized trial of MAMMS-IYCF against a standard of care in which nutritional screening and treatment is undertaken in accordance with Kenyan national guidelines by MOH clinicians. This study will not be blinded as the two-way mHealth system requires active patient engagement.

All children attending EID or HIV-care clinics in each health facility will be screened by the study staff for eligibility. Eligible children will be 6 months to 2 years of age (24 months) and exposed to HIV. Children with moderate or severe wasting (MUAC <12.5cm if older than 6-months, weight-for-height z-score <-2, or nutritional edema) at the time of screening will be excluded and referred for care. Written informed consent will be obtained in the caregivers chosen language.

After consent is provided, the child and caregiver's full socioeconomic and medical history will be taken including information on uptake of HIV specific managements. All enrolled children will participate in a physical examination performed by a study clinician trained in anthropometry, including dual measurement of length, weight, and MUAC, with a third measurement if there is >10% variability. Anthropometry of enrolled caregivers will also be done at enrollment, which includes height, weight and MUAC measurements. No biological samples will be collected during this study. A caregiver questionnaire will be administered including assessments of caregiver IYCF knowledge, trust in nutritional services, and intention to seek care if they believed their child to be wasted. Finally, all caregivers will be given basic nutrition education prior to randomization and will be trained on the use of MUAC tapes using a swap validation exercise in which caregivers will measure their child's MUAC and a study staff member will measure MUAC and the measurements will be compared for consistency. Training prior to randomization prevents biased exclusion of caregivers unable to pass swap validation.

Block randomization (1:1) with random sized blocks of no more than 12 will be developed by an independent biostatistician. Intervention arm allocation will be concealed in envelopes until the point of randomization. To ensure adequate numbers of HEU or CLWH in both arms, randomization will be stratified by known HIV status at enrollment using separate sets of envelopes for each status.

All caregivers randomized to the MAMMS-IYCF arm will be given two MUAC tapes after being trained by skilled study staff and their contact information will be entered into the online MAMMS-IYCF platform. The frequency of messages will be weekly. This will be a two-way SMS system and the content of each message was determined during the intervention pilot (KEMRI/SERU/CCR/309/4754), including health messages on age appropriate IYCF. These messages will also prompt caregivers to measure their child's MUAC, return the color outcome of the measurement to the clinic via the SMS system, and they will also be able to ask any health questions that they may have through the SMS. Separate automated messages will provide reminders of upcoming clinical or research visits that the caregiver and child require, to ensure good attendance at these visits. The caregivers can choose to receive these SMS in any of the four languages commonly spoken in the region: English, Swahili, and Luo. They can also choose the day and time messages are sent to them. The SMS messaging service will be free of cost to the caregivers as they will not incur any charges to send or receive messages while enrolled in the trial.

During follow-up, any caregiver who returns a MUAC measurement via SMS that suggests their child is wasted (yellow/red on the MUAC tape), will receive an SMS response asking them to come to the clinic with their child as soon as possible and this request will be followed up with a phone call. At this visit, the wasting status of the child will be confirmed by a study staff, and if treatment is indicated they will be enrolled into nutritional care provided by MOH nutritionists. If the caregiver incorrectly measured the MUAC, repeat training will be offered.

To promote high fidelity to current guidelines, all families randomized to the SOC arm will be enrolled into a one-way SMS service that will send them automated reminders before necessary routine EID or HIV care visits and research clinic appointments on day 90 and 180. During these routine clinic visits, study staff will ensure that the MUAC of SOC children are measured, and that treating clinicians (MOH nutritionists) are made aware if wasting is identified. Staff will ensure that SOC children are enrolled into nutritional care if they are wasted. This additional support means it is likely that SOC children in this study will receive care that is more adherent to current recommendations than is typical for many children. This enhanced SOC is necessary to ensure the investigators are comparing the MAMMS-IYCF to current guidelines. The SMS messaging service will be free of cost to the caregivers as they will not incur any charges to send or receive messages while enrolled in the trial.

Caregiver-child pairs will be asked to attend the research clinic at 90 and 180 days after enrollment to administer a follow-up CRF, take anthropometry measurements (height, weight and MUAC measurements) and collect recalled morbidity events. Where possible these visits will immediately follow routine clinical appointments, minimizing the travel burden on families. The investigators will also take anthropometry measurements of caregivers (weight, height and MUAC measurements) at 90 and 180 days after enrollment. At the final study visit, an SMS acceptability questionnaire will be administered to caregivers in the intervention arm to record any challenges or discomfort they felt using the system along with the pre/post-intervention secondary outcome survey questions. Any child noted to have malnutrition at the final study visit will be referred for treatment (a MUAC less than 12.5cm or any other common symptoms of malnutrition) and be administered with the acceptability questionnaire. During the study, caregivers will also be encouraged to come to the health facility for any medical concerns arising outside of the standard visit schedule.

The acceptability and feasibility of the MAMMS-IYCF intervention will be evaluated using a mixed-methods assessment of caregiver and nutritional program worker experiences and insights. In the RCT, all caregivers in the MAMMS IYCF arm will be administered an SMS acceptability questionnaire at study completion (180 days or the time at which a child is identified as malnourished), that will use Likert scores to quantify key indicators of participant satisfaction, tool acceptability, appropriateness, and feasibility. At completion of study, a subset of caregivers from the intervention arm across both study sites will be selected for focus groups discussions (FGD) or in-depth interviews if less than 15 participants are available to participant in each sampling group e.g. mothers of children with malnutrition and mothers of children who did not develop malnutrition. Focus group or in-depth interview participants will be randomly sampled from the pool of mothers, with 5-8 mothers sampled to participate in each FGD. If we do not have enough participants for the focus groups, we will conduct in-depth interviews with a subset of 5 participants in each of the following groups.

Two focus groups for each of the following sets of mothers will be conducted (possible range of 2-5, based on data saturation), with a minimum of eight focus groups in total. If less than 15 participants across both sites are available to participate from each sampling group, we will conduct in-depth interviews instead with 5 participants in each of the following groups (1) intervention arm participants whose child did not develop acute malnutrition, (2) intervention arm participants whose child developed acute malnutrition malnourished.

To complement, the investigators will conduct in depth interviews with mothers in the intervention arm to assess drivers of engagement with the platform. This includes about 10-16 caregivers who participated in the MAMMS intervention at varying levels of engagement.

The investigators will also conduct in-depth interviews with nutritional staff currently active in Migori and Homa Bay Counties and MAMMS-IYCF staff. Individual interviews will be conducted with purposively sampled health volunteers, including up to ten community health volunteers, three nurses, three nutritionists and three nutrition program managers (N=19 interviews with project implementers). The individual interviews will be used to understand health worker perceptions of the MAMMS-IYCF model, potential barriers to sustaining or scaling the program, and opportunities to improve the program moving forward. All FGDs and interviews will be conducted in the participant's preferred language (Kiswahili, Luo, or English) by a trained qualitative researcher and audio recorded for subsequent analysis. Interviews may be conducted in lieu of FGDs depending on the COVID-19 situation to respect social distancing recommendations.

Inclusion Criteria (HIV-exposed caregiver-child pairs):

• Children aged 6 to 24 months all-inclusive with a MUAC ≥ 12.5cm at the date of recruitment
• Children living with HIV or HIV-exposed uninfected children seen as outpatients in early infant detection (EID) or HIV-care clinics at the participating hospitals
• The child's caregiver is willing and able to provide informed consent
• The child's caregiver can read or write or has someone to help them read or write
• The child's caregiver is planning to remain in the catchment area with their child for > 6 months and willing to return to the health facility for 6-month follow up visits
• The child's caregiver has access to a Safaricom phone line and provides a mobile phone number

Inclusion Criteria (healthcare workers):

- Healthcare workers working in Homa Bay and Migori County Referral Hospitals, who have contact with pediatric inpatients

Exclusion Criteria:

• Children with moderate or severe wasting (MUAC <12.5cm, weight-for-height z-score <-2, or nutritional edema) at the time of eligibility screening
• Children with a congenital condition that limit feeding or syndromes that prevents age-appropriate feeding
• Child is enrolled in another study that the PI judges to compromise the aims of this study
• Child's caregiver does not pass the second training after being unable to satisfactorily complete the first MUAC training.
• Child's caregiver is under the age of 18 years.

De-identified IPD collected throughout the trial may be shared with other researchers within approximately 6 months of results publication